CN111866162A - A service distribution method and device - Google Patents

Abstract

The embodiment of the invention provides a service distribution method and device, relates to the technical field of communication, and can improve the rationality of service distribution. The method comprises the following steps: the service distribution equipment acquires computing power demand information of a target service; the method comprises the steps that the service distribution equipment determines at least two candidate computational force nodes from a plurality of computational force nodes, the at least two candidate computational force nodes meet a first preset condition, the first preset condition is that the connection time length of the at least two candidate computational force nodes and the service distribution equipment is smaller than a connection time length threshold value, and the credit degrees of the at least two candidate computational force nodes are larger than 0; the service distribution equipment determines the respective credibility of the at least two candidate computing power nodes; and the service distribution equipment determines the computing power node with the highest credibility in the at least two candidate computing power nodes as a target computing power node for processing the target service.

Description

A service distribution method and device

technical field

Embodiments of the present invention relate to the field of communications technologies, and in particular, to a service allocation method and apparatus.

Background technique

With the rapid development of communication technology, the data volume of data processing continues to grow, and the complexity of data processing is getting higher and higher. In an implementation manner, when a user equipment (user equipment, UE) processes a certain service (for example, image retrieval), the UE may send computing power requirement information to the service distribution device, and then the service distribution device determines to provide computing power for the UE The computing power node of the resource (hereinafter referred to as the target computing power node). Specifically, the service distribution device may determine the computing power node with the largest remaining computing power among the multiple computing power nodes as the target computing power node.

However, the above-mentioned method of determining the target computing power node only based on the remaining computing power may not meet other needs of the business. For example, the service's demand for delay cannot be met. In this way, the above-mentioned method for determining the target computing power node may not be reasonable enough.

SUMMARY OF THE INVENTION

Embodiments of the present invention provide a service allocation method and device, which can improve the rationality of service allocation. Further, the target computing power node determined in the embodiment of the present invention can satisfy the first preset condition and has high reliability, and using the target computing power node to process the target service can improve the success rate of processing the target service.

In a first aspect, an embodiment of the present invention provides a service distribution method, including: a service distribution device obtains computing power requirement information of a target service, where the computing power requirement information includes target computing power corresponding to the target service; At least two candidate computing power nodes are determined among the multiple computing power nodes, the remaining computing power of the multiple computing power nodes is greater than or equal to the target computing power corresponding to the target business, and the at least two candidate computing power nodes meet the first budget. Assuming a condition, the first preset condition is that the connection duration between the at least two candidate computing power nodes and the service distribution device is less than the connection duration threshold, and the credibility of the at least two candidate computing power nodes is greater than 0; the service distribution device Determine the respective reliability of the at least two candidate computing power nodes; the service distribution device determines the computing power node with the highest reliability among the at least two candidate computing power nodes as the target computing power node processing the target service.

In a second aspect, an embodiment of the present invention provides a service distribution device, including: an acquisition module and a determination module; the acquisition module is used to acquire computing power requirement information of a target service, and the computing power requirement information includes the corresponding data of the target service. target computing power; the determining module is used to determine at least two candidate computing power nodes from a plurality of computing power nodes, and the remaining computing power of the multiple computing power nodes is greater than or equal to the target computing power corresponding to the target business, the At least two candidate computing power nodes satisfy a first preset condition, and the first preset condition is that the connection duration between the at least two candidate computing power nodes and the service distribution device is less than the connection duration threshold, and the at least two candidate computing power The credibility of the node is greater than 0; and the respective credibility of the at least two candidate computing power nodes is determined; and the computing power node with the highest credibility among the at least two candidate computing power nodes is determined as the target of processing the target business Computing node.

In a third aspect, an embodiment of the present invention provides another service distribution device, including: a processor, a memory, a bus, and a communication interface; the memory is used to store computer execution instructions, and the processor and the memory are connected through a bus. When the service distribution device runs , the processor executes the computer-executable instructions stored in the memory, so that the service distribution apparatus executes the service distribution method provided in the first aspect.

In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, including instructions, which, when running on a service distribution apparatus, cause the service distribution apparatus to execute the service distribution method provided in the first aspect.

In a fifth aspect, an embodiment of the present invention provides a computer program product containing instructions, which, when the computer program product runs on a computer, enables the computer to perform the service allocation of the first aspect and any one of its implementations. method.

In the service distribution method and device provided by the embodiments of the present invention, the service distribution device obtains the computing power requirement information of the target service, and the computing power requirement information includes the target computing power corresponding to the target service; then, the service distribution device is based on the computing power node and The connection duration of the service distribution equipment and the reputation of the computing power node, from the multiple computing power nodes whose remaining computing power is greater than or equal to the target computing power, determine that the connection time with the service distribution equipment is less than the connection duration threshold, and the reputation degree is greater than or equal to the target computing power. 0 (that is, satisfying the first preset condition) at least two candidate computing power nodes; then the service distribution device determines the respective credibility of the at least two candidate computing power nodes, and the at least two candidate computing power nodes can be selected. The computing power node with the highest reliability is determined as the target computing power node processing the target business. In this embodiment of the present invention, because the service distribution device can determine whether the multiple computing power nodes are in a normal state and the reputation based on the respective connection durations between the multiple computing power nodes and the service distribution device and the respective reputations of the multiple computing power nodes at least two candidate computing power nodes with higher degrees, further, the service distribution device selects one of the at least two candidate computing power nodes with higher reliability in combination with the respective reliability of the at least two candidate computing power nodes The computing power node with higher reliability is determined as the target computing power node, which can improve the rationality of business distribution.

Further, the target computing power node determined in the embodiment of the present invention satisfies the above-mentioned first preset condition and has high reliability, and using the target computing power node to process the target service can improve the success rate of processing the target service.

Description of drawings

In order to illustrate the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that are required in the description of the embodiments or the prior art.

1 is a schematic diagram of an application scenario of a service allocation method provided by an embodiment of the present invention;

FIG. 2 is a schematic diagram of hardware of a server according to an embodiment of the present invention;

3 is a schematic diagram of a communication method provided by an embodiment of the present invention;

FIG. 4 is a schematic diagram of the internal structure of a computing power node according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram 1 of a service distribution apparatus according to an embodiment of the present invention;

FIG. 6 is a second schematic structural diagram of a service distribution apparatus according to an embodiment of the present invention.

Detailed ways

The service allocation method and apparatus provided by the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

Furthermore, references to the terms "comprising" and "having" in the description of this application, and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but optionally also includes other unlisted steps or units, or optionally also Include other steps or units inherent to these processes, methods, products or devices.

It should be noted that, in the embodiments of the present invention, words such as "exemplary" or "for example" are used to represent examples, illustrations, or descriptions. Any embodiments or designs described as "exemplary" or "such as" in the embodiments of the present invention should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "such as" is intended to present the related concepts in a specific manner.

The expression "and/or" in this application includes the use of either or both of the two methods.

In the description of this application, unless otherwise stated, the meaning of "plurality" refers to two or more.

Based on the problems existing in the background art, the embodiments of the present invention provide a service distribution method and device. The service distribution device obtains computing power requirement information of a target service, and the computing power requirement information includes the target computing power corresponding to the target service; then, the service Based on the connection time between the computing power node and the service distribution equipment and the reputation of the computing power node, the distribution equipment determines that the connection time with the service distribution equipment is shorter than the connection time from multiple computing power nodes whose remaining computing power is greater than or equal to the target computing power. At least two candidate computing power nodes with a duration threshold and a credibility greater than 0 (that is, satisfying the first preset condition); then the service distribution device determines the respective credibility of the at least two candidate computing power nodes, and assigns the at least two The computing power node with the highest reliability among the candidate computing power nodes is determined as the target computing power node processing the target business. In this embodiment of the present invention, because the service distribution device can determine whether the multiple computing power nodes are in a normal state and the reputation based on the respective connection durations between the multiple computing power nodes and the service distribution device and the respective reputations of the multiple computing power nodes at least two candidate computing power nodes with higher degrees, further, the service distribution device selects one of the at least two candidate computing power nodes with higher reliability in combination with the respective reliability of the at least two candidate computing power nodes The computing power node with higher reliability is determined as the target computing power node, which can improve the rationality of business distribution.

Further, the target computing power node determined in the embodiment of the present invention satisfies the above-mentioned first preset condition and has high reliability, and using the target computing power node to process the target service can improve the success rate of processing the target service.

A service distribution method provided by an embodiment of the present invention is applied to the service processing scenario shown in FIG. 1 . Specifically, the device in the service distribution system 10 (ie, the service distribution device 101 ) allocates appropriate computing power to the target service. node. Specifically, when the user (or user equipment) needs to process a certain service, the UE sends computing power requirement information to the service distribution device 101 in the service distribution system 10; when the service distribution device 101 obtains the computing power requirement information of the target service , the target computing node for processing the target service can be determined from the data processing system 20 . The data processing system 20 may include a plurality of computing power nodes, for example, including a computing power node 201 , a computing power node 202 , and a computing power node 203 . Among them, a computing power node can be composed of one or more devices. Exemplarily, as shown in FIG. 1 , the computing power node 201 includes device 2011, device 2012, and device 2013, the computing power node 202 includes device 2021 and device 2022, and the computing power node 203 includes device 2031, device 2032, and device 2022. 2033 and device 2034. Generally, in practical applications, the connections between the above-mentioned various devices or service functions may be wireless connections. In order to conveniently and intuitively represent the connection relationship between the various devices, solid lines are used in FIG. 1 for illustration.

Specifically, the computing power node 201 , the computing power node 202 , and the computing power node 203 all have a connection relationship with the service distribution device 101 . The computing power node 201, the computing power node 202 or the computing power node 203 may be used for processing services, that is, providing computing power corresponding to the service to the user (or UE). In this embodiment of the present invention, the service distribution device 101 may be based on the connection duration between the computing power nodes (including the computing power node 201, the computing power node 202, and the computing power node 203) and the service distribution device 101, the reputation of the computing power node, and the computing power The credibility of the node determines the target computing power node that processes the target business.

In this embodiment of the present invention, the computing power nodes included in the data processing system 20 may be one or more of terminal equipment, mobile edge computing (mobile edge computing, MEC) equipment, or data center equipment.

Optionally, the above data processing system may include one or more computing power nodes, and one computing power node may also include one or more devices. This embodiment of the present invention does not limit the number of each computing power node and device in the above-mentioned data processing system.

An embodiment of the present invention provides a service distribution device. The service distribution device may be a server. FIG. 2 is a schematic diagram of a hardware structure of a server that executes the service distribution method provided by the embodiment of the present invention. As shown in FIG. 2 , the server 30 may include A processor 301, a memory 302, a network interface 303, and the like.

The processor 301 is the core component of the server 30, and the processor 301 is used to run the operating system of the server 30 and the application programs (including system applications and third-party applications) on the server 30, so as to realize the service of the server 30. allocation method.

In this embodiment of the present invention, the processor 301 may be a central processing unit (central processing unit, CPU), a microprocessor, a digital signal processor (digital signal processor, DSP), an application-specific integrated circuit (application-specific integrated circuit, ASIC) ), field programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof, which can implement or perform various aspects described in conjunction with the disclosure in the embodiments of the present invention Exemplary logical blocks, modules and circuits; a processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

Optionally, the processor 301 of the server 30 includes one or more CPUs, and the CPU is a single-core CPU (single-CPU) or a multi-core CPU (multi-CPU).

The memory 302 includes, but is not limited to, random access memory (RAM), read only memory (ROM), erasable programmable read-only memory (EPROM), flash memory memory, or optical memory, etc. The code of the operating system is stored in the memory 302 .

Optionally, the processor 301 implements the service distribution method in the embodiment of the present invention by reading the instructions stored in the memory 302, or the processor 301 implements the service distribution method provided in the embodiment of the present invention by using internally stored instructions. When the processor 301 implements the service allocation method provided by the embodiment of the present invention by reading the execution stored in the memory, the memory stores an instruction for implementing the service allocation method provided by the embodiment of the present invention.

The network interface 303 is a wired interface, such as a fiber distributed data interface (FDDI) or a gigabit ethernet (GE) interface. Alternatively, the network interface 303 is a wireless interface. The network interface 303 is used by the server 30 to communicate with other devices.

The memory 302 is used to store the remaining computing power of the computing power node. Optionally, the memory 302 is further configured to store the connection duration between the computing power node and the service distribution device, and the like. The at least one processor 301 further executes the method described in the embodiment of the present invention according to the remaining computing power of the computing power node stored in the memory 302 and the connection duration between the computing power node and the service distribution device. For more details on how the processor 301 implements the above functions, please refer to the descriptions in the following method embodiments.

Optionally, the service distribution device further includes a bus, and the above-mentioned processor 301 and memory 302 are connected to each other through a bus 304, or are connected to each other in other ways.

Optionally, the service distribution device further includes an input-output interface 305, and the input-output interface 305 is used to connect with the input device and receive the computing power requirement information of the target service input by the user through the input device. Input devices include, but are not limited to, keyboards, touch screens, microphones, and the like. The input and output interface 305 is also used for connecting with an output device, and outputting the service distribution result of the processor 301 (that is, determining the target computing power node for processing the target service). Output devices include, but are not limited to, displays, printers, and the like.

In the embodiment of the present invention, since different services have different computing power requirements, when a user (or user equipment) needs to process a certain service (target service), the UE sends computing power requirement information to the service distribution device, so that the service distribution device The target computing power node for processing the target service may be determined based on information such as the computing power corresponding to the target service included in the computing power requirement information.

As shown in FIG. 3 , the service allocation method provided by the embodiment of the present invention may include S101-S104.

S101. The service distribution device acquires computing power requirement information of a target service.

The computing power requirement information of the target business includes the target computing power corresponding to the target business.

It should be understood that the target services include but are not limited to services such as image retrieval and image processing. The target computing power corresponding to the target business refers to the computing power that should be achieved by the computing power node processing the target business (ie, the target computing power node).

S102. The service distribution device determines at least two candidate computing power nodes from the plurality of computing power nodes.

Wherein, the remaining computing power of the multiple computing power nodes is greater than or equal to the target computing power corresponding to the target business. The at least two candidate computing power nodes satisfy a first preset condition, and the first preset condition is that the connection duration between the at least two candidate computing power nodes and the service distribution device is less than the connection duration threshold, and the at least two candidate computing power nodes The reputation of the node is greater than 0.

It should be understood that, after acquiring the target computing power corresponding to the target service, the service distribution device can select from the nodes included in the data processing system, multiple computing powers whose remaining computing power is greater than or equal to the target computing power (that is, the computing power requirement of the target service). power nodes, and then at least two candidate computing power nodes are determined from the plurality of computing power nodes.

In this embodiment of the present invention, before the above-mentioned S102 (or S101), the service distribution device may perform authentication and authentication on each computing power node (including multiple computing power nodes), so that each computing power node is respectively assigned to the service The device establishes a connection. Then, the service distribution device can distribute service to each computing power node.

It can be understood that after the above-mentioned computing power node and the service distribution device are connected, the service distribution device stores the connection status information of the computing power node, and the connection status information of the computing power node includes the identification of the computing power node, the computing power node and the service distribution equipment. The time when the connection was established and the current time. The following table 1 is an example of the connection status information of computing power nodes. In Table 1, the network access time of the computing power node is the time when the computing power node establishes a connection with the service distribution device.

Table 1

In this embodiment of the present invention, the above-mentioned computing power node connection status information is used to determine the connection time between the computing power node and the service distribution device, and the connection time between the computing power node and the service distribution device is the time between the computing power node and the service distribution device. The time difference of the current time. After the service distribution device determines the respective connection durations between the multiple computing power nodes and the service distribution device, the service distribution device determines, according to the preset connection duration threshold, candidate computing power nodes among the multiple computing power nodes whose connection duration is less than the connection duration threshold . For example, in Table 1, the five computing power nodes correspond to the same connection duration threshold, the connection duration threshold is 120 hours, and the service distribution device determines that the computing power node whose connection duration is less than the connection duration threshold is the computing power node 1, the computing power Node 2 and computing power node 3.

In this embodiment of the present invention, after the service distribution device determines the respective connection durations and connection duration thresholds between multiple computing power nodes and the service distribution device, it can determine candidate computing power nodes whose connection duration is less than the connection duration threshold.

Optionally, different computing power nodes may correspond to the same connection duration threshold, or may respectively correspond to different connection duration thresholds, which are not specifically limited in this embodiment of the present invention.

In the embodiment of the present invention, the reputation of a candidate computing power node satisfies:

C _n =C _an +C _bn -C _cn ;

Among them, C _n represents the reputation degree of the candidate computing power node, _{Can represents the initial reputation degree of the candidate computing power node, C bn represents} the corresponding gain reputation degree when the candidate computing power node successfully processes the business, and C _cn represents the candidate computing power node. The corresponding depreciation reputation when the computing power node fails to process the business.

It should be understood that each computing power node corresponds to a basic reputation degree when it leaves the factory (that is, after the production is completed), that is, the above-mentioned initial reputation degree. In the embodiment of the present invention, the initial reputation degree of each computing power node is the same. When a computing power node successfully processes a certain business, its credibility will be increased, and conversely, when the computing power node fails to process a certain business, its credibility will be correspondingly reduced.

Specifically, when a candidate computing power node successfully processes a business, the corresponding gain reputation degree satisfies:

Among them, C _bn represents the corresponding gain credibility when the candidate computing power node successfully processes the business, i represents that the candidate computing power node successfully processed i services within the preset time, and ω _j represents each of the i services. The corresponding computing power, ω _n represents the computing power of the candidate computing power node, and g is the gain factor.

In addition, when a candidate computing power node fails to process the business, the corresponding depreciation reputation degree satisfies:

Among them, C _cn represents the corresponding depreciation reputation when the candidate computing power node fails to process services, x represents that the candidate computing power node fails to process x services within the preset time, and ω _y represents each of the x services. The computing power corresponding to the business, ω _n represents the computing power of the candidate computing power node, and h is the depreciation factor.

Table 2 below is an example of the services successfully processed by the computing power node within the preset time and the services that failed to be processed within the preset time. As shown in Table 2, when i=0, it means that the computing power node has not successfully processed the business within the preset time, and when x=0, it means that the computing power node has not processed the failed service within the preset time.

Table 2

Based on Table 2, the service distribution device can determine the gain reputation degree and the deduction reputation degree corresponding to each of the five computing power nodes.

Exemplarily, it is assumed that the computing power (ie ω _n ) corresponding to the five computing power nodes shown in Table 2 is the same, which is 200 FLOP/S. It is also assumed that the initial credibility of the five computing power nodes is 100, g=40, h=20. Combined with the calculation formula of the reputation degree of the computing power node, the service distribution device can determine the reputation degree of each of the five computing power nodes, as shown in Table 3 below.

table 3

Computing node Reputation of computing power nodes Computing Node 1 100 Computing Node 2 70 Computing Node 3 -97.5 Computing Node 4 105 Computing Node 5 -1100

So far, the service distribution device determines that the computing power node 1, the computing power node 2, and the computing power node 4 are the computing power nodes whose reputation degree of the computing power nodes is greater than 0.

Combining the examples in Table 1 and Table 3, the service distribution device determines that among the five computing power nodes, computing power node 1 and computing power node 2 satisfy the first preset condition, that is, computing power node 1 and computing power node 2 are: Candidate computing power nodes.

Optionally, in another implementation manner, when the first computing power node does not meet the first preset condition, the service distribution device performs authentication on the first computing power node, so that the first computing power node Reconnect with the service distribution device.

The first computing power node is one of the above-mentioned multiple computing power nodes.

With reference to the description of the above-mentioned embodiments, it should be understood that when the connection duration between the first computing power node and the service distribution device is greater than or equal to the time threshold, it means that the connection duration between the first computing power node and the service distribution device is too long. Determine whether the first computing power node is in a normal state (or whether the first computing power node has security problems, including DOS attacks, password leakage, etc.). For example, when the first computing node is attacked by an attacker, it needs to continuously respond to the request packets sent by the attacker, which takes up a lot of memory. When the first computing node receives a real processing request, it may Unable to respond. Further, the service distribution device re-authenticates the first computing power node, so that the first computing power node and the service distribution device re-establish a connection; and/or, when the reputation of the first computing power node is less than or equal to 0 , indicating that the number of times when the first computing power node fails to process services within the preset time is greater than the number of times the first computing power node successfully processes services, because the performance of the equipment in the first computing power node, the resource utilization rate of the equipment is low, or the network Delay and other issues, so the first computing power node may be more likely to fail the business processing or the first computing power node can successfully process the business, but cannot complete the business processing within the preset time, and then the service distribution device can be used for the first computing power node. Re-authentication is performed, so that the first computing power node and the service distribution device are re-established.

It should be noted that, after the first computing power node and the service distribution device are re-established, the first computing power node can participate in the next service distribution. In addition, the service distribution device updates the stored computing power node connection status information, and updates the time when the computing power node and the service distribution device establish a connection to the current time (that is, the time when the connection is re-established), and the first computing power node has the same reputation. is equal to the initial reputation of the first computing power node.

In this embodiment of the present invention, because the service distribution device consumes resources every time the service distribution device performs authentication and authentication on the computing power node, the first computing power node is authenticated when the first computing power node does not meet the first preset condition. Authorization authentication can not only ensure the normal state of the first computing power node, improve the success rate of processing services, but also avoid the waste of resources caused by frequent authentication and authentication.

S103. The service distribution device determines the respective reliability of the at least two candidate computing power nodes.

It should be understood that the credibility of a candidate computing power node may be determined according to the computing power distributable rate of the candidate computing power node, the service quality of the candidate computing power node, the credibility of the candidate computing power node, and the like.

Taking candidate computing power node n as an example, in an implementation manner of the embodiment of the present invention, the credibility of a candidate computing power node satisfies:

Among them, Q _n represents the credibility of the candidate computing power node, rn represents the computing power distributable rate of the candidate computing power node, _{p n represents the service quality of the candidate computing power node, C n represents the} credibility of the candidate computing power node, C _an represents the initial credibility of the candidate computing power node, d, e, and f are constants, d+e+f=1.

Specifically, the computing power distributable rate of the candidate computing power node is the ratio of the remaining computing power of the candidate computing power node to the total computing power of the candidate computing power node, and the service quality of the candidate computing power node is the The product of the failure rate and the non-failure rate of the links contained in the candidate computing node.

It should be understood that the remaining computing power of the candidate computing power node is the difference between the total computing power of the candidate computing power node and the used computing power of the candidate computing power node. The computing power distributable rate of a candidate computing power node satisfies:

ω _{n remaining} = ω _n - ω _{n used}

Among them, rn represents the computing power distributable rate of the candidate computing power node, ω _n represents the total computing power of the candidate computing power node (that is, the computing power of the candidate computing power node), and the _{remainder of ω n represents} the computing power of the candidate computing power node. Remaining computing power, ω _{n used} indicates the used computing power of the candidate computing power node.

With reference to the description of the above embodiments, it should be understood that a computing power node may include at least one device (or server) of one or more devices, and the computing power of the computing power node is the sum of the computing power of each device in the one or one device .

Moreover, the quality of service of a computing power node is determined by the non-failure rate of one or more devices included in the computing power node and the non-failure rate of the link between the one or more devices.

Below, taking candidate computing power node n as an example, the computing power of candidate computing power node n satisfies:

ω _n =ω _n1 +ω _n2 +...+ω _nm

Among them, ω _n represents the computing power of the candidate computing power node n, m represents the number of devices included in the candidate computing power node n, and m is an integer greater than or equal to 1.

条链路，如此，该候选算力节点的服务质量满足：Moreover, for the candidate computing power node n, there is a connection relationship between any two devices inside it, so the candidate computing power node n exists inside

In this way, the service quality of the candidate computing power node satisfies:

表示该候选算力节点n中第

条链路的非故障率，m为大于或等于1的整数。Among them, p _n represents the service quality of the candidate computing power node n, p _nm represents the non-failure rate of the mth device in the candidate computing power node n,

Indicates the No. 1 in the candidate computing power node n

The non-failure rate of a link, m is an integer greater than or equal to 1.

Specifically, the non-failure rate of the mth device in the candidate computing node n satisfies:

P _nm = S _normal /S _working × P _{nm initial}

S _work = S _normal + S _fault

Among them, P _nm represents the non-failure rate of the m th device, S _normal represents the normal working time of the m th device, S _working represents the total working time of the m th device, and P _{nm initially} represents the m th device. The initial non-failure rate of , S _fault represents the faulty working time of the mth device (that is, the time when the mth device stops working normally due to the fault), and m is an integer greater than or equal to 1.

Similarly, the method for determining the non-failure rate of a link in the candidate computing power node n is similar to the method for determining the non-failure rate of a device in the candidate node n, and a link is used as an example below.

Specifically, the non-failure rate of the Mth link in the candidate computing node n satisfies:

l _nM = S' _normal / S' _working × l _{nM initial}

S' _working = S' _normal + S' _fault

Among them, l _nM represents the non-failure rate of the M-th link, S' _normal means the normal working time of the M-th link, S' _working means the total working time of the M-th link, and l _{nM initially} means The initial non-failure rate of the Mth link, S'failure represents the _faulty working time of the Mth link (that is, the time that the Mth link stops working normally due to the failure), and M is greater than or equal to Integer of 1, M=1,2,...,

It should be noted that when a candidate computing power node includes only one device, the service quality of the candidate computing power node is independent of the link, and the service quality of the candidate computing power node is the non-failure rate of the device.

Combined with the example in S102 above, it is assumed that the remaining computing power of the computing power node 1 is 200 FLOP/S, and the remaining computing power of the computing power node 2 is 180 FLOP/S. Then, the service distribution device determines that the computing power distributable rate of the computing power node 1 is 100%, and the computing power distributable rate r ₂ of the computing power node 2 is 90%.

Exemplarily, as shown in Figure 4, computing power node 1 includes 2 devices and 1 link, namely device 11, device 12 and link ①; computing power node 2 includes 3 devices and 3 links , namely device 21, device 22, device 23, link ②, link ③, and link ④. Assume that the non-failure rates of the two devices included in the computing power node 1 shown in the figure are the same as the non-failure rates of the three devices included in the computing power node 2, both of which are 99%; The non-failure rate of one link included is the same as the respective non-failure rates of the three links included in Hash Node 2, both of which are 98%. Then the service distribution device determines that the service quality of the computing power node 1 is 96.05%, and the service quality p ₂ of the computing power node 2 is 94.12%.

From the examples in Table 2 above, it can be seen that the initial reputation of computing power node 1 and computing power node 2 are the same, both are 100, the reputation degree of computing power node 1 (ie C ₁ ) is 100, and the reputation of computing power node 2 is The degree (ie C ₂ ) is 70.

It is also assumed that d=0.4, e=0.1, and f=0.5. In this way, the service distribution device determines that the reliability of the computing power node 1 is 0.996, and the reliability of the computing power node 2 is 0.811.

S104: The service distribution device determines the computing power node with the highest reliability among the at least two candidate computing power nodes as the target computing power node for processing the target service.

Exemplarily, with reference to the example in S103 above, the service distribution device determines the computing power node 1 as the target computing power node for processing the target service.

Embodiments of the present invention provide a service distribution method and device. The service distribution device obtains computing power requirement information of a target service, and the computing power requirement information includes target computing power corresponding to the target service; then, the service distribution device is based on the computing power node and the computing power. The connection duration of the service distribution equipment and the reputation of the computing power node, from the multiple computing power nodes whose remaining computing power is greater than or equal to the target computing power, determine that the connection time with the service distribution equipment is less than the connection duration threshold, and the reputation degree is greater than or equal to the target computing power. 0 (that is, satisfying the first preset condition) at least two candidate computing power nodes; then the service distribution device determines the respective credibility of the at least two candidate computing power nodes, and the at least two candidate computing power nodes can be selected. The computing power node with the highest reliability is determined as the target computing power node processing the target business. In this embodiment of the present invention, because the service distribution device can determine whether the multiple computing power nodes are in a normal state and the reputation based on the respective connection durations between the multiple computing power nodes and the service distribution device and the respective reputations of the multiple computing power nodes at least two candidate computing power nodes with higher degrees, further, the service distribution device selects one of the at least two candidate computing power nodes with higher reliability in combination with the respective reliability of the at least two candidate computing power nodes The computing power node with higher reliability is determined as the target computing power node, which can improve the rationality of business distribution.

Further, the target computing power node determined in the embodiment of the present invention satisfies the above-mentioned first preset condition and has high reliability, and using the target computing power node to process the target service can improve the success rate of processing the target service.

In this embodiment of the present invention, the service distribution device and the like can be divided into functional modules according to the foregoing method examples. For example, each functional module can be divided corresponding to each function, or two or more functions can be integrated into one processing module. The above-mentioned integrated modules can be implemented in the form of hardware, and can also be implemented in the form of software function modules. It should be noted that, the division of modules in the embodiment of the present invention is schematic, and is only a logical function division, and there may be other division manners in actual implementation.

In the case where each functional module is divided according to each function, FIG. 5 shows a possible schematic structural diagram of the service distribution apparatus involved in the above embodiment. As shown in FIG. 5 , the service distribution apparatus 40 may include: obtaining module 401 and determination module 402.

The obtaining module 401 is configured to obtain computing power demand information of a target business, where the computing power demand information includes target computing power corresponding to the target business.

A determination module 402, configured to determine at least two candidate computing power nodes from a plurality of computing power nodes, the remaining computing power of the multiple computing power nodes is greater than or equal to the target computing power corresponding to the target business, the at least two candidate computing power nodes The computing power node satisfies a first preset condition, and the first preset condition is that the connection duration between the at least two candidate computing power nodes and the service distribution device is less than the connection duration threshold, and the reputation of the at least two candidate computing power nodes is and determine the respective reliability of the at least two candidate computing power nodes; and determine the computing power node with the highest reliability among the at least two candidate computing power nodes as the target computing power node processing the target service.

Optionally, the reputation of a candidate computing power node satisfies:

C _n =C _an +C _bn -C _cn

Among them, C _n represents the reputation degree of the candidate computing power node, _{Can represents the initial reputation degree of the candidate computing power node, C bn represents} the corresponding gain reputation degree when the candidate computing power node successfully processes the business, and C _cn represents the candidate computing power node. The corresponding depreciation reputation when the computing power node fails to process the business.

Optionally, the credibility of a candidate computing power node satisfies:

Among them, Q _n represents the credibility of the candidate computing power node, rn represents the computing power distributable rate of the candidate computing power node, _{p n represents the service quality of the candidate computing power node, and C n represents the} candidate computing power node. The credibility of , _Can represents the initial credibility of the candidate computing power node, d, e, f are constants, d+e+f=1. The computing power distributable rate of the candidate computing power node is the ratio of the remaining computing power of the candidate computing power node to the total computing power of the candidate computing power node, and the service quality of the candidate computing power node is the equipment included in the candidate computing power node. The product of the non-failure rate of , and the non-failure rate of the links contained in the candidate computing node.

Optionally, the service distribution apparatus 40 further includes a storage module 403 .

The storage module 403 is used to store the connection status information of the computing power node, the connection status information of the computing power node includes the identifier of the computing power node, the time when the computing power node established the connection with the service distribution device, and the current time, and the computing power node is connected. The state information is used to determine the connection duration between the computing power node and the service distribution device.

Optionally, the service distribution apparatus 40 further includes an authentication and authentication module 404 .

The authentication and authentication module 404 is configured to perform authentication and authentication on the first computing power node when the first computing power node does not meet the first preset condition, so that the first computing power node is allocated with the service The device re-establishes the connection, and the first computing power node is one of the plurality of computing power nodes.

In the case of using an integrated unit, FIG. 6 shows a possible schematic structural diagram of the service distribution apparatus involved in the above embodiment. As shown in FIG. 6 , the service distribution apparatus 50 may include: a processing module 501 and a communication module 502 . The processing module 501 may be used to control and manage the actions of the service distribution apparatus 50. For example, the processing module 501 may be used to support the service distribution apparatus 50 to perform S102, S103 and S104 in the above method embodiments. The communication module 502 may be used to support the communication between the service distribution apparatus 50 and other entities. For example, the communication module 502 may be used to support the service distribution apparatus 50 to perform S101 in the foregoing method embodiments. Optionally, as shown in FIG. 6 , the service distribution apparatus 50 may further include a storage module 503 for storing program codes and data of the service distribution apparatus 50 .

Wherein, the processing module 501 may be a processor or a controller (for example, it may be the above-mentioned processor 301 shown in FIG. 2 ). The communication module 502 may be a transceiver, a transceiver circuit, or a communication interface, etc. (for example, it may be the above-mentioned network interface 303 shown in FIG. 2 ). The storage module 503 may be a memory (eg, may be the memory 302 shown in FIG. 2 described above).

Wherein, when the processing module 501 is a processor, the communication module 502 is a transceiver, and the storage module 503 is a memory, the processor, the transceiver and the memory can be connected through a bus. The bus may be a peripheral component interconnect (PCI) bus, an extended industry standard architecture (EISA) bus, or the like. The bus can be divided into address bus, data bus, control bus and so on.

It should be understood that, in various embodiments of the present invention, the size of the sequence numbers of the above-mentioned processes does not mean the sequence of execution, and the execution sequence of each process should be determined by its functions and internal logic, rather than the embodiments of the present invention. implementation constitutes any limitation.

Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of the present invention.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described systems, devices and units may refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

The units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In the above-mentioned embodiments, it may be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using a software program, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in accordance with the embodiments of the present invention are generated in whole or in part. The computer may be a general purpose computer, special purpose computer, computer network, or other programmable device. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server, or data center Transmission to another website site, computer, server, or data center is performed by wire (eg, coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (eg, infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or data storage devices including one or more servers, data centers, etc. that can be integrated with the medium. The usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVD), or semiconductor media (eg, Solid State Disk (SSD)), and the like.

The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention. should be included within the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims (10)

1. A method for service allocation, comprising:

the method comprises the steps that service distribution equipment obtains calculation force demand information of a target service, wherein the calculation force demand information comprises target calculation force corresponding to the target service;

the service distribution equipment determines at least two candidate calculation force nodes from a plurality of calculation force nodes, wherein the residual calculation forces of the calculation force nodes are all larger than or equal to the target calculation force corresponding to the target service, the at least two candidate calculation force nodes meet a first preset condition, the first preset condition is that the connection time length of the at least two candidate calculation force nodes and the service distribution equipment is smaller than a connection time length threshold value, and the credit degrees of the at least two candidate calculation force nodes are larger than 0;

the service distribution equipment determines the respective credibility of the at least two candidate computational power nodes;

and the service distribution equipment determines the calculation power node with the highest credibility in the at least two candidate calculation power nodes as a target calculation power node for processing the target service.

2. The method of claim 1, wherein the reputation of a candidate force node satisfies:

C_n＝C_an+C_bn-C_cn；

wherein, C_nRepresenting the reputation of the candidate force node, C_anRepresenting an initial reputation, C, of the candidate force node_bnRepresenting the corresponding gain credit degree when the candidate calculation force node successfully processes the service, C_cnAnd representing the corresponding profit reduction credit degree when the candidate calculation force node fails to process the service.

3. The method of claim 2, wherein the confidence level of a candidate force node satisfies:

wherein Q is_nRepresenting the confidence level, r, of the candidate computational power node_nComputing power assignability ratio, p, representing said candidate computing power nodes_nRepresenting the quality of service of the candidate computational power node, C_nRepresenting the reputation of the candidate force node, C_anRepresenting the initial credibility of the candidate calculation force node, wherein d, e and f are constants, and d + e + f is 1; the computing power allocable rate of the candidate computing power node is the ratio of the residual computing power of the candidate computing power node to the total computing power of the candidate computing power node, and the service quality of the candidate computing power node is the product of the non-failure rate of the equipment contained in the candidate computing power node and the non-failure rate of the link contained in the candidate computing power node.

4. The method according to any one of claims 1 to 3, wherein the service distribution device stores therein computational power node connection state information, the computational power node connection state information includes an identifier of a computational power node, a time at which the computational power node establishes a connection with the service distribution device, and a current time, and the computational power node connection state information is used to determine a connection duration of the computational power node with the service distribution device.

5. The method according to any one of claims 1 to 3, further comprising:

and under the condition that a first computational power node does not meet the first preset condition, the service distribution equipment performs authentication and authentication on the first computational power node so as to reestablish connection between the first computational power node and the service distribution equipment, wherein the first computational power node is one of the computational power nodes.

6. A traffic distribution apparatus, comprising: the device comprises an acquisition module and a determination module;

the acquisition module is used for acquiring computing power demand information of a target service, wherein the computing power demand information comprises target computing power corresponding to the target service;

the determining module is configured to determine at least two candidate computation force nodes from a plurality of computation force nodes, where remaining computation forces of the computation force nodes are all greater than or equal to a target computation force corresponding to the target service, the at least two candidate computation force nodes satisfy a first preset condition, and the first preset condition is that connection time lengths of the at least two candidate computation force nodes and the service distribution device are less than a connection time length threshold, and reputation degrees of the at least two candidate computation force nodes are greater than 0; and determining respective trustworthiness of the at least two candidate force nodes; and determining the computing power node with the highest credibility in the at least two candidate computing power nodes as a target computing power node for processing the target service.

7. The traffic distribution apparatus of claim 6, wherein the reputation of a candidate computation force node satisfies:

C_n＝C_an+C_bn-C_cn；

wherein, C_nRepresenting the reputation of the candidate force node, C_anRepresenting an initial reputation, C, of the candidate force node_bnRepresenting the corresponding gain credit degree when the candidate calculation force node successfully processes the service, C_cnAnd representing the corresponding profit reduction credit degree when the candidate calculation force node fails to process the service.

8. The traffic distribution apparatus of claim 7, wherein the confidence level of a candidate computational power node satisfies:

wherein Q is_nRepresenting the confidence level, r, of the candidate computational power node_nComputing power assignability ratio, p, representing said candidate computing power nodes_nRepresenting the quality of service of the candidate computational power node, C_nRepresenting the reputation of the candidate force node, C_anRepresenting the initial credibility of the candidate calculation force node, wherein d, e and f are constants, and d + e + f is 1; the computing power allocable rate of the candidate computing power node is the ratio of the residual computing power of the candidate computing power node to the total computing power of the candidate computing power node, and the service quality of the candidate computing power node is the product of the non-failure rate of the equipment contained in the candidate computing power node and the non-failure rate of the link contained in the candidate computing power node.

9. A service distribution arrangement according to any of claims 6 to 8, further comprising a memory module;

the storage module is used for storing the connection state information of the computational power node, the connection state information of the computational power node comprises an identifier of the computational power node, the time for establishing connection between the computational power node and the service distribution device and the current time, and the connection state information of the computational power node is used for determining the connection duration between the computational power node and the service distribution device.

10. A service distribution device according to any one of claims 6 to 8, further comprising an authentication module;

the authentication module is configured to authenticate a first force node when the first force node does not satisfy the first preset condition, so that the first force node is reestablished from the service distribution apparatus, where the first force node is one of the plurality of force nodes.

Images